Anticoagulation prescription in atrial fibrillation

Aims: We seek to assess the factors associated with the anticoagulation prescription in a cohort of patients with atrial fibrillation (AF) collected from out-patient clinics. Methods: A total of 1524 patients with a history of AF were collected from out-patients clinics. CHADS2, CHA2DS2-VASc and HAS-BLED scores were calculated in every patient. Variables associated with anticoagulant treatment prescription were analyzed in univariant and multivariant models. Results: Most patients received either anticoagulant (62%) or antiplatelet treatment (37%). Anticoagulation rates increased among higher CHADS2 and CHA2DS2-VASc score values. A logistic regression model was performed to assess the variables associated with the prescription of anticoagulant treatment; the variables with stronger association were the presence of arrhythmia at the current visit (odds ratio (OR) 33, 95% CI 27 - 40, p < 0.001) and lack of concomitant antiplatelet treatment (OR 0.17, 95% CI 0.14 - 0.21, p < 0.001). Conclusions: Although prognosis of patients with AF is mainly determined by the long-term thrombotic risk, the prescription of antithrombotic therapy depends more on the bleeding risk and the immediate thrombotic risk perception.